optical motional feedback

The optical system is just method of providing a signal that represents the actual motion of the speaker cone.
This signal would be fed back into a servo type amplifier and used to correct the cone motion to reduce distortion.
The speaker and amplifier become a "closed-loop" system, which in theory, would produce more accurate sound compared to the traditional open-loop system.
The patent only covers the method of creating the feedback signal by using light, as opposed to other methods like piezoelectric or electromagnetic transducers.
It doesn't mention the purpose of the feedback signal or the overall concept of a closed-loop system.

The reason light can be used to detect motion is because light intensity varies with distance (inverse square law). In this case the light source is constant intensity from an LED which is mounted on the moving speaker cone. A photo transistor detects the light, which is now slightly variable intensity due to the cone motion and produces an electrical signal that is proportional to the cone motion.
 
As far as i know, Backes&Müller (Germany) had several speakers with cone movement control via capacity sensing. Inside their first model ->BM5 (or Monitor 5) on every transducer, in subsequent models only within the (Super-)Tweeters (Models BM6, BM12, etc.).
Some years ago there was some information on the net. I have not checked, if these are still available.

Eberhard
 
The reason light can be used to detect motion is because light intensity varies with distance (inverse square law). In this case the light source is constant intensity from an LED which is mounted on the moving speaker cone. A photo transistor detects the light, which is now slightly variable intensity due to the cone motion and produces an electrical signal that is proportional to the cone motion.
this was just what I wanted to know.
so it doesn't use a laser beam.
why is there no mention of this system? it does not work well ?
 
why is there no mention of this system? it does not work well ?
The reason light can be used to detect motion is because light intensity varies with distance (inverse square law). In this case the light source is constant intensity from an LED which is mounted on the moving speaker cone. A photo transistor detects the light, which is now slightly variable intensity due to the cone motion and produces an electrical signal that is proportional to the cone motion.
I would think it is very hard to obtain a good signal-to-noise ration from such a sensing method. At higher frequencies and lowel levels, the movement is truly microscopic and the light density modulation now is so small that it's often down in the noise... and we have a number of noise sources here.

The position sensing principle generally suffers from the 2nd order roll-off of the signal, for each doubling of signal frequency the output reduces to a quarter, plus the servo-loop must take care of the double integration (after applying a linearization).

Maybe with a whole array of sensors the noise and precision would be good enough for subwoofer and woofer application. Multiple sensing locations also should even out the impact of cone-breakup when strategically selected. And because its response goes down to DC, a control loop can be very effective for woofers, even tackling dynamic DC offsets and other strange ill-behaviors of woofers.

Sadly, optical solutions need a good shielding of ambient and stray light, a very precise alignment, and are prone to degrade from dust and an dirt, besides electrical aging effects, the problem of likely needing to linearize the signal, etc.
Overall, optical motional feedback for speakers is generally difficult to implement and limited in application.
 
  • Like
Reactions: 1 users
As far as i know, Backes&Müller (Germany) had several speakers with cone movement control via capacity sensing. Inside their first model ->BM5 (or Monitor 5) on every transducer, in subsequent models only within the (Super-)Tweeters (Models BM6, BM12, etc.).
Some years ago there was some information on the net. I have not checked, if these are still available.

Eberhard
I believe B&M also used an optical system, consisting of a conical pin connected to the cone, and a light detecting circuit (light source and a LDR). The pin moved in and out the light detecting circuit, blocking more or less light. The position of the cone thus could be read. Differentiating this gave the cone speed. IIRC.
 
what do you think about linear magnetic sensors? .
there are those with hall effect or mageresistive which includes three subfamilies.
could it be used to detect the difference in distance of the magnetic field generated by the permanent magnet of the same speaker?
 
I'm sure there are many sensing methods that could work, including even a simple microphone.
I've often thought of trying a dual voice coil subwoofer by driving it with one coil and sensing with the other.
I suppose there would be some signal coupling between the coils since they are probably wound together.
A differential amp may be needed to extract the motion signal from the coupled drive signal.
It may not be the best approach but I think it could be a good starting point for a DIY experiment.

The whole idea of correcting cone motion is intriguing but I really don't know effective it can be since I've never had the chance to hear one.
I think sensing is the simpler part of the process and the creation / tuning / application of the corrective signal is far more difficult to get right, considering the dynamic nature of music. I do have some experience with servo motors and industrial motion control in my job. Of course this is much simpler than controlling a speaker cone, but the principles are the same. However, even these can be quite challenging to tune for reliable performance under all conditions.
 
Member
Joined 2007
Paid Member
Along the microphone idea powersoft M-force can use a differential pressure sensor with a pressure sensor on each side of the cone.

high speed laser interferometry could work optically and I believe is used as part of driver development to measure excursion but the equipment costs more than a transducer. I was able to use a cheap IR reflection based sensor to measure excursion but the sample rate is low so it only worked for repetitive signals: https://data-bass.ipbhost.com/topic/760-measuring-excursion/?do=findComment&comment=19480

Practicaly its usualy cheaper to improve woofer linearity by using a more capable woofer than adding feedback to a woofer.
 
Over here we have quite a few DIY folks that continue the tradition Philips started with it's piezo motion detector. It's easier to board that ship than to venture on your own, as the complications in the electronic feedback mechanism are a bit tricky (the oscillation risks particularly, but getting things linear too).